Update app.py
Browse filesworking for 1 url
app.py
CHANGED
|
@@ -118,20 +118,14 @@ def classify_website(url):
|
|
| 118 |
|
| 119 |
|
| 120 |
urls = [url]
|
| 121 |
-
print(urls)
|
| 122 |
results_shop = main(urls)
|
| 123 |
|
| 124 |
# Convert results to DataFrame
|
| 125 |
df_result_train_more = pd.DataFrame(results_shop)
|
| 126 |
-
print(df_result_train_more)
|
| 127 |
text = df_result_train_more['text'][0]
|
| 128 |
-
|
| 129 |
-
try:
|
| 130 |
-
translated = GoogleTranslator(source='auto', target='en').translate(text[:4990])
|
| 131 |
-
# except:
|
| 132 |
-
|
| 133 |
|
| 134 |
-
|
| 135 |
# Prepare the input prompt for the model
|
| 136 |
prompt = f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
|
| 137 |
### Instruction:
|
|
@@ -161,38 +155,14 @@ Categorize the website into one of the 3 categories:
|
|
| 161 |
logging.exception(e)
|
| 162 |
return str(e)
|
| 163 |
|
| 164 |
-
|
| 165 |
-
def classify_urls_from_csv(csv_file):
|
| 166 |
-
# Read CSV file and extract URLs from the first column
|
| 167 |
-
df = pd.read_csv(csv_file)
|
| 168 |
-
df.iloc[:, 0] = df.iloc[:, 0].str.replace(';', '').str.strip()
|
| 169 |
-
|
| 170 |
-
urls = df.iloc[:, 0].tolist()
|
| 171 |
-
|
| 172 |
-
# Classify each URL and store the results
|
| 173 |
-
predictions = []
|
| 174 |
-
for url in urls:
|
| 175 |
-
prediction = classify_website(url)
|
| 176 |
-
predictions.append(prediction)
|
| 177 |
-
|
| 178 |
-
# Add predictions as a new column in the dataframe
|
| 179 |
-
df['Prediction'] = predictions
|
| 180 |
-
|
| 181 |
-
# Save the results to a new CSV file
|
| 182 |
-
output_file = "predictions.csv"
|
| 183 |
-
df.to_csv(output_file, index=False)
|
| 184 |
-
|
| 185 |
-
return output_file
|
| 186 |
-
|
| 187 |
-
|
| 188 |
# Create a Gradio interface
|
| 189 |
iface = gr.Interface(
|
| 190 |
-
fn=
|
| 191 |
-
inputs=
|
| 192 |
-
outputs=
|
| 193 |
title="Website Categorization",
|
| 194 |
-
description="
|
| 195 |
)
|
| 196 |
|
| 197 |
# Launch the interface
|
| 198 |
-
iface.launch()
|
|
|
|
| 118 |
|
| 119 |
|
| 120 |
urls = [url]
|
|
|
|
| 121 |
results_shop = main(urls)
|
| 122 |
|
| 123 |
# Convert results to DataFrame
|
| 124 |
df_result_train_more = pd.DataFrame(results_shop)
|
|
|
|
| 125 |
text = df_result_train_more['text'][0]
|
| 126 |
+
translated = GoogleTranslator(source='auto', target='en').translate(text[:4990])
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
|
| 128 |
+
try:
|
| 129 |
# Prepare the input prompt for the model
|
| 130 |
prompt = f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
|
| 131 |
### Instruction:
|
|
|
|
| 155 |
logging.exception(e)
|
| 156 |
return str(e)
|
| 157 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
# Create a Gradio interface
|
| 159 |
iface = gr.Interface(
|
| 160 |
+
fn=classify_website,
|
| 161 |
+
inputs="text",
|
| 162 |
+
outputs="text",
|
| 163 |
title="Website Categorization",
|
| 164 |
+
description="Categorize a website into one of the 3 categories: OTHER, NEWS/BLOG, or E-commerce."
|
| 165 |
)
|
| 166 |
|
| 167 |
# Launch the interface
|
| 168 |
+
iface.launch()
|